geekhack
geekhack Community => Keyboards => Topic started by: ZX Spectrum on Tue, 08 February 2022, 07:57:46
-
Recently I have bought a brand new IBM Model M (1391401) made by Lexmark 1994.
when I tried to connect the keyboard I have noticed that not all key are registering. besides, caps-lock is not turning on.
below is a screenshot from Switch Hitter and key log as well.
notice the strange pattern of faulty switches... does anyone recognize a know issue with the matrix that would cause such behavior?
-
Those outages follow the horizontal traces. First step would be to open it up and reseat the ribbon connectors.
-
Those outages follow the horizontal traces. First step would be to open it up and reseat the ribbon connectors.
actually I haven't opened it up yet.
I will do that and get back to you
-
Those outages follow the horizontal traces. First step would be to open it up and reseat the ribbon connectors.
well here is the pictures of the ribbon connectors. I have noticed a black spots on the traces but on the back side of the ribbon.
do you think these indicate parts of traces are worn off?
-
Plug it back together and see if things work.
-
Plug it back together and see if things work.
no it doesn't work. exactly the same problem remains
-
Plug it back together and see if things work.
no it doesn't work. exactly the same problem remains
There's an old trick for cleaning oxidized contacts. Try rubbing them gently with a pencil eraser until they look clean.
-
If you do the eraser thing be VERY gentle. I'd advise against doing that at all.
Your issue with the skinnier ribbon cable (the one for the rows). Next step would be to plug it all back together, press down on a key that's unresponsive, and push on the ribbon cable, just deforming it a bit back and forth. With the mark on it I'd more suspect a break in the trace than an issue with how dirty the tabs are.
You can see the matrix for an ssk in this thread (which is the same minus the numpad of course) https://geekhack.org/index.php?topic=59027.0
-
If you do the eraser thing be VERY gentle. I'd advise against doing that at all.
I have done it hundreds, maybe thousands, of times with ribbon cables in laptop computers with no ill effect. If the trace is broken, he's got bigger problems than damaging the contacts. They're typically very robust if they're not shorting and burning up (seen plenty of that in laptop video cables).
That being said, you would know more specifically about a 1994 Lexmark keyboard membrane ribbon cable than I would.
-
If you do the eraser thing be VERY gentle. I'd advise against doing that at all.
I have done it hundreds, maybe thousands, of times with ribbon cables in laptop computers with no ill effect. If the trace is broken, he's got bigger problems than damaging the contacts. They're typically very robust if they're not shorting and burning up (seen plenty of that in laptop video cables).
That being said, you would know more specifically about a 1994 Lexmark keyboard membrane ribbon cable than I would.
don't you think the whole problem is just broken traces ??
don't these black spots means that ??
-
don't you think the whole problem is just broken traces ??
Those traces are definitely no longer continuous.
But while I do not have personal experience repairing them, I do know that other people have been successful with it.
-
don't you think the whole problem is just broken traces ??
Those traces are definitely no longer continuous.
But while I do not have personal experience repairing them, I do know that other people have been successful with it.
I think there are special type of pens that you can use to draw conducting traces
do you know what are they called or where I can find them?
-
I haven't used these, so someone else might recommend better, but something like this:
https://www.ebay.com/itm/275165229654?hash=item4011203256:g:MHQAAOSwksNiB~Wz (https://www.ebay.com/itm/275165229654?hash=item4011203256:g:MHQAAOSwksNiB~Wz)
-
The problem here is the protective cover, the traces aren't exposed there, and you'd have to get under it.
It looks like the break is right on the transition. I've had this happen and what I did was jam a very thin wire (a strand from a multistrand wire) under the cover and extending out a little, then taping it down. Not my proudest fix, but did work.
-
I haven't used these, so someone else might recommend better, but something like this:
https://www.ebay.com/itm/275165229654?hash=item4011203256:g:MHQAAOSwksNiB~Wz (https://www.ebay.com/itm/275165229654?hash=item4011203256:g:MHQAAOSwksNiB~Wz)
thank you for the information
-
The problem here is the protective cover, the traces aren't exposed there, and you'd have to get under it.
It looks like the break is right on the transition. I've had this happen and what I did was jam a very thin wire (a strand from a multistrand wire) under the cover and extending out a little, then taping it down. Not my proudest fix, but did work.
what do you mean by protective cover??
actually I think I am not quit getting your idea
-
The problem here is the protective cover, the traces aren't exposed there, and you'd have to get under it.
It looks like the break is right on the transition. I've had this happen and what I did was jam a very thin wire (a strand from a multistrand wire) under the cover and extending out a little, then taping it down. Not my proudest fix, but did work.
what do you mean by protective cover??
actually I think I am not quit getting your idea
I think he's saying that there are multiple layers of plastic and the traces themselves are not exposed, but sandwiched between. This is typical with membrane traces. They're only exposed where they connect to the controller.
I have seen traces that look like that that work perfectly fine (I deal with student Chromebooks, so I see one heck of a lot of damage that should literally never happen). I have seen traces that are even partially burnt up from liquid damage that still have continuity and work fine. Kind of hard to test with a multimeter if the damaged parts of the traces aren't exposed unless you can follow them back to another point they are exposed on the membrane and test from where they connect to the controller, or if the protective layer can be lifted enough nondestructively.
I haven't used conductive ink, etc. I have read of problems with getting a proper electrical connection with them, excessive resistance (maybe not a problem here) and their being relatively fragile. It may not hold up well in places the membrane flexes. I haven't used it, but I have ordered some of this MG Chemicals conductive epoxy (https://www.amazon.com/gp/product/B08K3TG5D9/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&th=1) because my research suggested that it was more resilient and less resistive than other alternatives. They're a trusted brand in soldering supplies as well. I was going to try to use it to replace the factory LEDs on a Mini M membrane but I haven't gotten around to hacking it up.
E TwentyNine's use of thin wire sounds the least destructive. Definitely worth trying to narrow down the problem to that specific place if a continuity tester/multimeter won't work as easily.
-
Recently I have bought a brand new IBM Model M (1391401) made by Lexmark 1994.
when I tried to connect the keyboard I have noticed that not all key are registering. besides, caps-lock is not turning on...
Recently I have bought a brand new IBM Model M (1391401) made by Lexmark 1994.
when I tried to connect the keyboard I have noticed that not all key are registering. besides, caps-lock is not turning on...
First, dude, I'm sorry you're having trouble with this board. Assuming you didn't just luck onto it somewhere (e.g. a dark, creepy abandoned warehouse full of forgotten BS keebs—which would be where, exactly, BTW?), it couldn't have been cheap. Either way, it must've been quite disappointing to find that a board so renowned for its reliability (it wasn't made in Scotland, right? LOL) didn't work straight out of the box. Really, what are the odds?
If you don't manage to fix the cable(s), I wonder if Unicomp has some that'd work? A 1994 M, made only a year or two before the Unicomp guys took over the Lexmark factory (https://en.wikipedia.org/wiki/Unicomp), would be as architecturally similar to a present-day Unicomp M as possible. This may have already occurred to you, but I thought I'd mention it. Cheers!
-
Recently I have bought a brand new IBM Model M (1391401) made by Lexmark 1994.
when I tried to connect the keyboard I have noticed that not all key are registering. besides, caps-lock is not turning on...
Recently I have bought a brand new IBM Model M (1391401) made by Lexmark 1994.
when I tried to connect the keyboard I have noticed that not all key are registering. besides, caps-lock is not turning on...
First, dude, I'm sorry you're having trouble with this board. Assuming you didn't just luck onto it somewhere (e.g. a dark, creepy abandoned warehouse full of forgotten BS keebs—which would be where, exactly, BTW?), it couldn't have been cheap. Either way, it must've been quite disappointing to find that a board so renowned for its reliability (it wasn't made in Scotland, right? LOL) didn't work straight out of the box. Really, what are the odds?
If you don't manage to fix the cable(s), I wonder if Unicomp has some that'd work? A 1994 M, made only a year or two before the Unicomp guys took over the Lexmark factory (https://en.wikipedia.org/wiki/Unicomp), would be as architecturally similar to a present-day Unicomp M as possible. This may have already occurred to you, but I thought I'd mention it. Cheers!
actually it was by Lexmark In us as I mentioned.
I just wanted to figure out the problem before I actually order a replacement membrane from Unicomp and go for bolt modding it (which I have no experience or tools for )
-
Recently I have bought a brand new IBM Model M (1391401) made by Lexmark 1994.
when I tried to connect the keyboard I have noticed that not all key are registering. besides, caps-lock is not turning on...
First, dude, I'm sorry you're having trouble with this board. Assuming you didn't just luck onto it somewhere (e.g. a dark, creepy abandoned warehouse full of forgotten BS keebs—which would be where, exactly, BTW?), it couldn't have been cheap. Either way, it must've been quite disappointing to find that a board so renowned for its reliability (it wasn't made in Scotland, right? LOL) didn't work straight out of the box. Really, what are the odds?
If you don't manage to fix the cable(s), I wonder if Unicomp has some that'd work? A 1994 M, made only a year or two before the Unicomp guys took over the Lexmark factory (https://en.wikipedia.org/wiki/Unicomp), would be as architecturally similar to a present-day Unicomp M as possible. This may have already occurred to you, but I thought I'd mention it. Cheers!
actually it was by Lexmark In us as I mentioned.
I just wanted to figure out the problem before I actually order a replacement membrane from Unicomp and go for bolt modding it (which I have no experience or tools for )
I think all you really need is a drill and some screwdrivers and/or nut drivers (depending on method), don't you? Definitely a last resort either way.
I did something similar to my cheaply-made ABKO Hacker (https://geekhack.org/index.php?topic=115217.0) as well. Not high up on the difficulty scale as far as keyboard modding goes.
-
a brand new IBM Model M
not all key are registering
a dark, creepy abandoned warehouse
"new" and "spent the last 26 years in a pleasant, climate-controlled environment" are not necessarily the same thing.
You should probably check for corrosion or other electrical problems first.
-
"new" and "spent the last 26 years in a pleasant, climate-controlled environment" are not necessarily the same thing...
Well, there you go, eh? Maybe NOS should mean "Not Opened, So...?"
(I wish I could've spent the last 26 years in a pleasant environment... But I'm married.)